Magnetic field effects on crystallization by LLIP method

I. Yamamoto1, T. Okabe1, M. Tatara1, Y. Chiba1, T. Onotou1, and N. Hirota21 Yokohama National University, Tokiwadai, Hodogayaku, Yokohama 240-8501, Japan

2 National Institute for Materials Science, Tsukuba 305-0047, Japan

Application of magnetic fields to crystallization have been studied in order to make a high quality and large crystal.1-3)

Liquid-liquid interfacial precipitation (LLIP) method is one of technique of crystallization from solution. Two kinds of solvent are stacked to make their interface. One is a poor solvent for desired material, and the other is a good solvent saturated with the material. In general, right solvent is stacked on heavy one to prevent the convection owing to gravity. The seed crystal is born and grown at around the two-dimensional interface because the supersaturated layer is generated due to mutual diffusion of the two liquid. The sedimentation speed is low for small crystal and accelerated with increasing size of crystal according to Stroke’s low. When the grown crystal is leaved from the interface and sunk to the bottom then the growth reaction is completed. Magnetic field and its gradient are thought to be influenced the crystallization processesas follows. The generation of seed crystal and the growth rate are controlled by the magnetic field because the degree of the super-saturation depended on the diffusion of solution is controlled by Lorentz force under the influence of magnetic field. The growth direction is also controllable magnetically because the posture of crystal is controllable against the two-dimensional interface if the crystal has anisotropic magnetic susceptibility. In addition, the crystal size iscontrollable because the staying period at around the super-saturated layer, which corresponds the reaction period, is controllable according to apply of magnetic force parallel or antiparallel to gravity.In the experiments, some kinds of crystal were crystallized and the magnetic field effects of the size, morphology, magnetic orientation, and quality of crystalwere estimated for C60-fullerene nano-rod (FNR), NaCl salt, ice, glycine, taurine, lysozyme, thaumatin, etc. as listed in Table 1.4) For example, the long axis of FNRwas oriented perpendicular to the magnetic flux as shown in Fig. 1(b). The volume of FNR was enlarged by 10times in the homogeneous horizontal magnetic field of 13 T. Under the influence of the gradient vertical magnetic field, the volume was enlarged 100 times in the reduced gravity environment as shown in Fig. 2. The size effects were also recognized for other crystals as listed in Table 1. Moreover, changes of crystal hobbit and morphologywere found as a magnetic field effect.The magnetic field effect on size must be applicable to crystallization of protein for drag discovery, sincecrystallization for many unknown proteins is hard to make a good large crystal. If huge crystal is precipitated then the protein is analyzed easily by XRD structure analysis. A hen egg white lysozyme was crystalized by the LLIP method under the influence of magnetic field of up to 13 T. Ten times huge protein crystal was observed and its XRD structure analysis showed that the crystal kept high quality with the maximum resolution of 1.22 Åand R-merge of 4.6% (Fig. 3). The high mosaicity was

(a) B = 0 T (b) Horizontal B = 9.6 T (c) Vertical B = 7.2 T

Fig. 1 SEM images of FNRs crystalized in (a) zeromagnetic fields, (b) horizontal homogeneous field of 9.6 T, and (c) vertical field of B = 7.2 T with gradient of dB/dz= −58 T/m. The directions of magnetic flux were shown as allow in images (b) and (c). The scale bars indicated 1 μmfor (a) and (b), and 10 μm for (c).

Fig. 2 Scatter plot for fullerene nano rod crystallized under the influence of magnetic fields. The symbols ●, ×,and , corresponded to the size for Fig. 1(a), (b) and (c), respectively.

11pA 1

1

expected because the crystal aligned perfectly to magnetic flux.In addition, a new one-dimensional reaction field was developed among a paramagnetic liquid and two kinds of diamagnetic liquid under the influence of horizontal gradient magnetic field as shown in Fig. 4. The paramagnetic liquid was forced toward the magnetic center (the left-hand direction in Fig. 4) due to horizontal magnetic force. The horizontal two-dimensional interface was changed vertical by the high magnetic field and the coexistent line among three liquidswas appeared. We except the new reaction field is available to make new low dimensional products.

Table 1. Typical experimental condition of LLIP method and the magnetic field effect

Raw material/ Good solvent

Poor solvent(precipitant)

Diameter of reactor, d / mm

Reaction duration

Magneticfield effect

Remarks

C60 fullerene/ Toluene

2-propanol 14.5 48 hr Size x100,Orientation

rod crystal

Ice (water)/ 1-BuOH

Toluene 5.0 3 hr Root position,Orientation,

Growth rate x6

20 oC

Glycine/ Water

EtOH 14 2 hr Polymorph,Orientation

0 oC

Taurine/ Water

EtOH 14 2 hr Crystal habit,Orientation

Thaumatin/ Acid

(Rochelle salt) 6.8 20 day Orientation pH = 5.0

Lysozyme/ TAE

PEG4000, (CoCl2, etc.)

6.8 5 day ~ 2 month

Size x10, Orientation

negative g

Acknowledgements: This work was partially supported by JSPS KAKENHI Grant Number 16K04946 and JASRI/Spring-8 Grant Number 2016B2899. Experiments were performed at Instrumental Analysis Center of Yokohama National University and NIMS. The SAXS was measured at KEK-PF and BL38B1 in SPring-8, Japan.

References1) S. Chandrasekhar, Philos. Mag. 43 (1952) pp. 501-532.2) H. P. Utech and M. C. Flanagans, J. Appl. Phys. 37 (1966) pp. 2021-2024.3) H. A. Chedzy and D. T. Hurle, Nature 210 (1966) pp. 933-934.4) T. Onotou and I. Yamamoto, Proc. Materials Analysis and Processing in Magnetic Fields (2016).

Fig. 3. The structure of lysozyme crystal precipitated under magnetic field with BdB/dz = −587 T2/m.

Fig. 4 Side view of the vertical interface under horizontal gradient magnetic field.the interface was indicated as broken line.

2

Numerical simulation on structure formation of magnetic particles under magnetic fields

T. Ando1, D. Katayama1, N. Hirota2, O. Koike3, R. Tatsumi4 and M. Yamato5 1 College of Industrial Technology, Nihon University, Narashino 275-8575, Japan

2 Fine Particle Engineering Group, National Institute for Materials Science, Tsukuba 305-0047, Japan 3 Products Innovation Association, Tokyo 113-8656, Japan

4 Environmental Science Center, The University of Tokyo, Tokyo 113-8656, Japan 5 Graduate School of Urban Environmental Sciences, Tokyo Metropolitan University, Hachioji 192-0397, Japan

This study gives an attention to structure formation of magnetic particles by dipole-dipole interaction under magnetic fields in materials processing. When magnetic field is applied to medium containing magnetic particles, the direction of magnetic moment of magnetic particles align in the same direction as the magnetic field, and magnetic particles make chain clusters by the dipole interaction among particles1. So, the magnetic field is useful tool on development of anisotropic materials. In this time, we performed numerical simulations of the structure formation of magnetic particles whose size is on the order of micrometers in material processing, and studied the unsteady process of the formation and its feature. Figure 1 shows the schematic model of numerical simulation. The cube of simulation region is a part of the inside of the container and the periodic boundary condition for a 3D system is used. Magnetic particles are randomly dispersed in the container at the initial condition. The translational motion and the rotational motion of particles are governed by Newton and Euler equation, respectively. Here, we used DEM including the magnetic effect. About the simulation research of magnetic particles, A. Satoh has already published many studies2. He deals with magnetic particles in magnetic fluids, and particle size is on the order of tens of nanometers. Therefore, he calculates using Brownian dynamics method in which particles do not collide directly with each other. He also discusses mainly the rheology of magnetic fluid due to the structure of magnetic particle, and not structure formation in unsteady process. One example of numerical simulation results is shown in Fig. 2. Particles aligned in the direction of the magnetic field and made the structure formation of chain cluster. We discuss the structure formed by the magnetic particles and the formation process when the concentration and particle size of the particles in dispersion medium are changed.

Reference 1) M. Yamato, et al., Polymer, 55 (2014) 6546-6551. 2) A. Satoh, Introduction to Molecular-Microsimulation of Colloidal Dispersions, Elsevier, (2003).

Fig. 1 Model of numerical simulation

(a) Isometric view

Fig. 2 Result of numerical simulation

(b) Perpendicular section to the direction of magnetic field

11pA 2

3

Liquid Crystal Magneto-Electropolymerization

Hiromasa Goto

Department of Materials Science, Faculty of Pure and Applied Materials Science, University of Tsukuba

1. Alignment of Liquid Crystal Conjugated Polymers Magnetic field [1] or shear stress [2] affords to produce uniaxial

alignment of liquid crystals. Orientation of liquid crystal conjugated

polymer has been achieved under magnetic field. The aligned polymer

thus obtained in the magnetic field shows main-chain orientation

accompanied by orientation of liquid crystal substituents.

2. Liquid crystal electropolymerization Electrochemical polymerization in liquid crystals has been carried

out. The polymer obtained in liquid crystal matrix shows liquid crystal

like morphology observable with optical microscopy [3] and scanning

electron microscopy. Although the polymer shows liquid crystal like

optical texture, the polymer shows no fluidity (polymer solid film). This

is due to the fact that molecular collective form imprinting from liquid

crystal matrix to resultant polymer in the polymerization process was

occurred.

3. Liquid crystal chiral electropolymerization Chiral conjugated polymers were prepared by electrochemical

polymerization of achiral monomers in a chiral liquid crystal (CLC)

electrolyte solution [1,2]. The polymer films prepared in chiral liquid

crystal shows “Electro-driven change in optical rotation”. The optical

rotation degree is

comparable to that of

Faraday rotators. The

optical rotation degree

can be precisely

controlled be the external

voltage less that 1 V.

This can be a new

physical effect in optical

rotators. The ellipticity of

this polymer is also

found to exhibit

hysteresis.

4. Liquid Crystal Magneto-Electrochemical Polymerization We carried out electrochemical synthesis of conducting polymers

in magnetic field of 12 T. The polymer thus obtained in the oriented

liquid crystal show uniaxial form. Polarized absorption spectra of the

polymers confirmed anisotropy and "linear polarized electrochromism

[2]. The uniaxial optical function is controlled by application of external

voltage.

5. Liquid crystal magnetic orientation in solvent evaporation process We developed a new method of magnetic orientation in solvent evaporation process via LC state to obtain aligned

polymer. Uniaxial alignment of conjugated polymer in liquid crystal was achieved under magnetic field. A growth of

the liquid crystal domains and magnetic orientation occur simultaneously in this process to form thin solid films with

align liquid crystal order.

Reference

[1] Goto, H. Phys. Rev. Lett. 2007, 98, 253901.

[2] Goto, H; Nimori, S. J. Mater. Chem., 2010, 20, 1891–1898

Figure 1. Polarizing optical microscopy

images of unoriented polymer (top) and

oriented polymer with magnetic field

(bottom).

Figure 2. Alignment of a liquid crystal

conjugated polymer with shear stress.

Figure 3. A polymer prepared in liquid

crystal under magnetic field.

H

H

11pA 3

4

Composite Coatings Utilizing Magnetically Fixed Particles

Junji Sasano, Takuo Ebitani, Takamichi Yamamoto, Seiji Yokoyama, and Masanobu Izaki Toyohashi University of Technology, Toyohashi 441-8580, Japan

Composite coating is one of the methods to form films of composite materials utilizing electrodeposition. The electrodeposited film with co-deposited particles is augmented with additional functions which cannot be possessed by the matrix metal film alone. In general, composite coatings are formed by electrodepositing the matrix metal films co-deposited with particles from their suspended solutions. Homogeneity and amount of particles in the films, and the stability of suspended solutions depend on the characteristics of the particles and the matrix metals, so it is not easy to find the optimum condition to control these properties simultaneously. To overcome this problem, we have investigated a new method to form similar composite structures without using suspended solutions; the particles mixed with magnetic particles are fixed on the electrode by the attractive force from the magnets placed on its back side, and then, their gaps are filled with electrodeposited metal. Here, we call this method as composite coatings utilizing magnetically fixed particles (CCMFP). In this paper, we investigate alumina co-deposited nickel composite coatings as a model case of CCMFP, and show some necessity conditions to realize the formation of composite coatings by this method.

As a preliminary test, we tried to find the suitable condition for filling the gaps between nickel particles magnetically fixed on a copper working electrode with electrodeposited nickel without using alumina particles. Conventional Watts bath, which is one of the simplest nickel plating solution, was used as the electrolyte solution. The magnets for fixing the nickel particles were two types of neodymium magnets, we call them magnet A and B; the magnet A has the magnetic flux density of 360 mT with the size of 15×10×5 cm3 and the magnet B is a bundle of small magnets having the magnetic flux density of 213 mT with the size of 1×1×5 cm3 (total size: 16×14×5 cm3). (Fig. 1) Electrodeposition using magnet A, however, did not result in the formation of a uniform film. One reason was that nickel was electrodeposited mainly on the nickel particles not on the copper electrode due to the conductivity of nickel particles. This problem was overcome by increasing the contact resistivity between each particle with thermally-formed nickel oxide. Another reason was non-uniform distribution of the magnetic field, and that was evaded using the magnet B with a bundle structure. After improving these two issues, bottom-up growth of electrodeposited nickel was achieved, but even so, the roughness of the film did not improve. We thought this was because the bottom-up filling of nickel slowed down during the electrodeposition process due to the recovery of conductivity between nickel particles by the dissolution of nickel oxide. To investigate the stability of the oxide, we performed electrodeposition using solutions with different pH values; then electrochemical measurements and thermodynamic estimations revealed that the nickel oxide was easily reduced at low pH, and moderately high pH without forming precipitates of nickel hydroxide in the solution was suitable for continuing uniform bottom-up filling. Nickel-alumina composite coating was conducted under the optimized condition written above, and it was successfully formed. (Fig. 2) Although the surface roughness of the composite film and uniformity of dispersion of alumina particles need further improvement, the results demonstrate the possibility of CCMFP as a potent method for the formation of composite coatings.

Fig. 1 Experimental setup of the three-electrode electrochemical cell for CCMFP with a Cu plate working electrode with magnetically fixed Ni particles, a Ni wire quasi reference and a Ni counter electrode. Photos of the magnet A and B, and the schematic structure of the magnet B are also presented.

Fig. 2 Cross-sectional SEM image of Al2O3-Ni composite film formed by CCMFP; darker gray parts are included Al2O3 particles.

100 μm

WECERE

++

Magnet

Cu working electrode with magneticallyfixed Ni particles

Cu plateNi particles

Nd magnet1×1×5 mm3

N

S

NSNS…16 mm

Magnet A Magnet B

11pA 4

5

High magnetic field effect on copper electrodeposition and anodic dissolution

○Y. Oshikiri1, M. Miura2, S. Takagi3 and T. Asada3 1Yamagata College of Industry and Technology, Yamagata 990-2473, Japan.

2Hokkaido Polytechnic College, Otaru, Hokkaido 047-0292, Japan. 3Fuculty of Symbiotic Systems Science, Fukushima University, Fukushima 960-1296, Japan.

Magnetoelectrochemistry is the electrochemistry in magnetic field, which is characterized by a macroscopic

solution flow called magnetohydrodynamic (MHD) flow induced by Lorentz force resulting from electrolytic current and magnetic field. Mass transport in magnetic field is strongly affected by MHD flow, so that in electrodeposition and anodic dissolution, surface morphology of electrode can be controlled by magnetic field.

Numerous microscopic flows called micro MHD flow spontaneously occur in MHD flow. In a vertical magnetic field, a macroscopic tornado-like vortex called vertical MHD flow emerges over an electrode called vertical MHD electrode (VMHDE). In electrodeposition and anodic dissolution, lots of vortexes of micro MHD flows create screw dislocations having chirality depending on the direction of magnetic field. According to this process, copper electrodes with high chiral catalytic activities have been fabricated and applied to enantiomeric reactions [1]. However, in general, because of viscosity, about 1 μm vortexes mentioned above are impossible to rotate. For such a high performance of catalysis, something to drastically decrease the viscosity is required. One question therefore arises; what is it?

Answer is ionic vacancy created as a byproduct in electrode reaction; a polarized spherical free space of the order of 0.1 nm surrounded by oppositely charged ionic cloud, which is produced by the conservations of linear momentum and electric charge during electron transfer [2]. The average lifetime is about 1 sec [3], which is extraordinarily long comparing with a collision period of 10-10 sec of solution particle. The most important point is its migration without entropy production, which implies that it does not interplay with other solution particles, providing inviscid flow without viscosity. In copper anodic dissolution on a copper VMHDE, microbubbles originated from ionic vacancies has been observed (Fig. 1) [4]. Furthermore, the micro MHD flows make ionic vacancies distributed in mosaic, so that on the patterned vacancy layer, drilling effect of micro MHD vortexes leads to characteristic pit formation with high aspect ratio on a copper electrode surface (Magneto-drill effect) (Fig. 2).

Electrodeposition is also strongly affected by magnetic field. Unrelated with hydrogen evolution, copper dendrites develop in high magnetic fields (Magneto-dendrite effect) (Fig. 3) [5], where in place of hydrogen molecules, nanobubbles from ionic vacancies adsorb on 3D copper nuclei, inducing remarkable dendritic growth.

Reference 1) I. Mogi, R. Aogaki, & K. Watanabe: Sci. Rep. 3, 2574 (2013). 2) R. Aogaki, et al: Sci. Rep., 6, 28927 (2016). 3) A. Sugiyama, et al: Sci. Rep., 6, 19795 (2016). 4) Y. Oshikiri, et al: Electrochemistry, 83, 549 (2015). 5) M. Miura, et al: Sci. Rep., 7, 45511 (2017).

Fig. 3 Magneto-dendrite effect in copper deposition. (a) B = 15 T without hydrogen gas evolution. (b) B = 0 T with hydrogen gas evolution.

Fig. 2 Magneto-drill effect on a copper rod electrode at 10T.

Fig 1. Microbubble globules on copper surface during copper dissolution at 8 T.

Microbubble

1mm 5µm

11pA 5

6

Concept and Procedure for the Synthesis of Uniform Nanoparticles in Liquid Phase with Large Quantity

A. Muramatsu Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan

Since monodispersed particles have excellent uniformity in size, shape, composition, and structure both of bulk and surface, they can be used as highly functional materials for industrial use due to unified characteristics under one standard based on their monodispersity. Among so many synthesis method of nanomaterials, Gel-Sol Method1) is unique in the production scheme from the dissolution of initially formed gel to the re-precipitation of monodispersed particles in solution phase, where the composition of ions and/or complexes are kept constant in the solution phase. For this method, particles are formed through nucleation and growth by a direct deposition of monomers onto growing particles, obeying LaMer model2). Namely, the growth mechanism of particles is not via aggregation of primary particles. For hydrothermal processes, including Gel-Sol Method, the most important technique is to avoid tremendous coagulation between growing particles. As you know, DLVO theory3) predicts adequate conditions of particle synthesis that we usually choose dilute solution system, where electrolyte concentration is too low to lead to the rapid aggregation. In contrast, Gel-Sol system shows the concentrated solution, including initially formed gel. For this method, growing particles are strongly immobilized in very viscous gel network to completely reduce Brownian motion, a trigger for the aggregation, in which a particle collides with another one. Namely, the particle formation rate is preponderantly higher than the coagulation one of growing particles in order to prevent aggregation. In this manner, Gel-Sol method can be applied in non-aqueous solution in place of aqueous one so as to establish the combination of solvothermal synthesis with Gel-Sol method. For hydrothermal procedure, we have been experienced that materials, which is predicted stoichiometrically to form in any condition, sometimes cannot be prepared. For example, α-Al2O3 cannot be formed by the hydrothermal synthesis method, where AlOOH, intermediate compound between initially formed Al(OH)3 and α-Al2O3 as a final product, is too stable to dissolve in aqueous solution in order to convert into α-Al2O3. For the other example, ITO, indium tin oxide, is the same case: ITO is often used as Transparent Conductive Film on various devices, such as LCD, SmartPhone, etc. Although ITO is expected to form stoichiometrically via a hydrothermal method, ITO does not be obtained in the actual synthesis process. For these cases, we choose organic solvent such as alcohol etc., in

Na0.5K0.5NbO3

TiO2

TiO2

Fe2O3 Fe2O3 Fe2O3

ITO

ITO

11pA 6

7

Fe(acac)3 2.0 mmolOleic acid 5.0 mLOleylamine 5.0 mL

Heated at 320 oC

Washed with ethanol

FeO NPs

N2 atmosphere

PreheatingHeated at 210 oC

Dispersed in Hexane

40.9 ± 3.8 nm

FeO

23.6 ± 2.0 nm 59.5 ± 8.2 nm

■ The particle size could be controlled by changing the preheating time.

The number of nucleus were increased by long time heating, thus the final particle size were decreased with time.

■ Cubic shaped FeO nanoparticles were formed.

Stirred for 15 min

FeO NPs 360 mgCyclohexane 6 mL

Tetraethyl OrthosilicateFe/Si = 0.1 [mol / mol]

Stirred for 16 h

Washed with MeOH, EtOHFeO@ SiO2 NPs

Cyclohexane 80 mLIGEPAL CO-520 12 mL25% NH3 aq. 2 mL

XRF measurementFe/Si = 0.11

Fe/Si = 0.13

Fe/Si = 0.11

Uniform SiO2 shell was formed on the surface of FeO nanoparticles.

SiO2

FeO

place of water, since there is too much oxygen sources, that is, water. Namely, the synthesis of magnetic nanoparticles is exemplified4). It is very difficult to obtain single-phase ε-Fe2O3 nanoparticles, because ε-Fe2O3 phase is an intermediate phase on the transformation from γ-Fe2O3 phase to α-Fe2O3 phase. To prepare ε-Fe2O3 phase selectively, focusing on the formation energy of ε-Fe2O3 phase is important. The actual preparation method is shown as follows. First, FeO nanoparticles with the size of 24, 40 and 60 nm were synthesized by our previous method4). As-obtained nanoparticles were coated with SiO2 shell to prevent their tremendous aggregation during heat treatment. The resulting SiO2 coated FeO nanoparticles were treated at 1000 - 1200 °C in the air. When 40 nm FeO nanoparticles with ca. 15 nm SiO2 shell were treated, significant aggregation and increase in particle size were observed at every treatment temperatures. As-prepared nanoparticles were found the mixture of γ-Fe2O3, ε-Fe2O3 and α-Fe2O3 phases. When 40 nm FeO nanoparticles were treated to cover with ca. 40 nm SiO2 shell, the particle aggregation was completely inhibited. The crystal structure was found to transform from γ-Fe2O3 phase to α-Fe2O3 phase via ε-Fe2O3 phase as the treatment temperature increased. The ε-Fe2O3 nanoparticles formed at 1100 °C showed large coercivity of 20.8 kOe at 300 K. As a result, the control in change in the crystal structure with temperature, under the complete inhibition against aggregation nor the growth in size, is the most important so that ε-Fe2O3 is selectively formed as a desired product.

Reference 1) T. Sugimoto, M. M. Khan, A. Muramatsu, H. Itoh, Colloids Surf. A, 79, 233 (1993). T. Sugimoto, Monodispersed

Particles, Elsevier, Amsterdam, pp. 376-388, (2001). 2) V. K. LaMer, Ind. Eng. Chem., 44, 1270–1277 (1952). 3) B. Derjaguin and L. Landau, "Theory of the stability of strongly charged lyophobic sols and of the adhesion of

strongly charged particles in solutions of electrolytes", Acta Physico Chemica URSS, 14: 633 (1941), E. J. W. Verwey and J. Th. G. Overbeek, Theory of the stability of lyophobic colloids, Amsterdam: Elsevier. (1948).

4) M. Nakaya, R. Nishida, N. Hosoda, A. Muramatsu, Crystal Research Tech., 52, 11 (2017). 5) M. Nakaya, R. Nishida, A. Muramatsu, Chem. Lett. 42, 863 (2013).

8

11pA-111pA-211pA-311pA-411pA-511pA-6